refactor: adopt 36 upstream cleanups with Pylon adaptations

[rynfar]

Obsolete helpers and tests of their implementation details make it harder to maintain the active Pylon paths. Adopt 36 reviewed upstream cleanup commits, remove unused code, and exercise retained behavior through the public APIs used by the product.

Adapt conflicts to Pylon's provider selection and capacity retention, Cursor discovery cache, held-send recovery, preview controls, and branding. Preserve draft annotation recovery from #330. The theme editor already uses the vivid palette generator; its semantic contrast coverage remains. Keep the existing pricing lookup equivalent when removing its normalization wrapper. No dependency, protocol, persistence version, layout, or animation changes.

The upstream review ledger lists every source SHA and three concrete deferrals for code Pylon still uses. All source patches and PR context were reviewed before adoption. Grok discovery cleanup was re-evaluated and adopted after #310 merged; DEF-16 is retired. The complete upstream cursor remains unchanged.

Validation: 1,125 focused tests in 36 files passed, including the corrected pricing adaptation. Shared, contracts, client-runtime, server, web, mobile, and desktop typechecks passed; lint and formatting passed on all 66 surviving changed code/test files. After rebasing onto the merged skill-discovery changes, 89 provider tests and the server typecheck passed. Real worker and Pylon-specific regression tests remain. Native-device testing and deployment are not claimed.

Model: GPT-6 Astra. Harness: Codex.

[github-actions]

Thread transfer impact

✅ Thread transfer remains within every enforced ceiling.

Provider	Metric	Main baseline	This PR	Impact	PR ceiling
Codex	Total thread wire	13.7 KiB	13.6 KiB	−37 B (−0.3%)	15.1 KiB	✅
Codex	Thread snapshot wire	7.1 KiB	7.1 KiB	−17 B (−0.2%)	7.3 KiB	✅
Codex	Live turn WebSocket wire	6.6 KiB	6.5 KiB	−20 B (−0.3%)	7.8 KiB	✅
Codex	Live turn WebSocket decoded	56.5 KiB	56.5 KiB	0 B (0.0%)	66.4 KiB	✅
Codex	Live turn messages	8	8	0 (0.0%)	21	✅
Claude	Total thread wire	13.7 KiB	13.8 KiB	+156 B (+1.1%)	15.1 KiB	✅
Claude	Thread snapshot wire	7.1 KiB	7.1 KiB	−12 B (−0.2%)	7.3 KiB	✅
Claude	Live turn WebSocket wire	6.6 KiB	6.7 KiB	+168 B (+2.5%)	7.8 KiB	✅
Claude	Live turn WebSocket decoded	57.3 KiB	58.2 KiB	+928 B (+1.6%)	66.4 KiB	✅
Claude	Live turn messages	8	10	+2 (+25.0%)	21	✅

Baseline: 72db913 · PR result: 996867d · Source CI: success

Scenario and decoded snapshot size

10 historical turns, 5 command tools per turn, 878.9 KiB retained MCP result per historical turn, and a 1.05 MiB retained result in the measured turn.

• Codex decoded thread snapshot: 111.2 KiB
• Claude decoded thread snapshot: 111.9 KiB

Updated in place by a trusted workflow. PR artifacts are strictly validated and never executed.